Polyketide synthase (PKS) systems are generally known in the art as enzyme complexes derived from fatty acid synthase (FAS) systems, but which are often highly modified to produce specialized products that typically show little resemblance to fatty acids. Researchers have attempted to exploit polyketide synthase (PKS) systems that have been described in the literature as falling into one of three basic types, typically referred to as: Type II, Type I iterative (fungal-like) and Type I modular. The Type II system is characterized by separable proteins, each of which carries out a distinct enzymatic reaction. The enzymes work in concert to produce the end product and each individual enzyme of the system typically participates several times in the production of the end product. This type of system operates in a manner analogous to the fatty acid synthase (FAS) systems found in plants and bacteria. Type I iterative (fungal-like) PKS systems are similar to the Type II system in that the enzymes are used in an iterative fashion to produce the end product. In other words, the same set of reactions is carried out in each cycle until the end product is obtained. There is no allowance for the introduction of unique reactions during the biosynthetic procedure. The Type I iterative (fungal-like) PKS system differs from Type II in that enzymatic activities, instead of being associated with separable proteins, occur as domains of larger proteins. This system is analogous to the Type I FAS systems found in animals and fungi.
In contrast to the Type I iterative (fungal-like) and Type II systems, in Type I modular PKS systems, each enzyme domain is used only once in the production of the end product. The domains are found in very large proteins that do not utilize the economy of iterative reactions (i.e., a distinct domain is required for each reaction) and the product of each reaction is passed on to another domain in the PKS protein. Additionally, in all of the PKS systems described above, if a carbon-carbon double bond is incorporated into the end product, it is always in the trans configuration.
Polyunsaturated fatty acids (PUFAs) are critical components of membrane lipids in most eukaryotes (Lauritzen et al., Prog. Lipid Res. 40 1 (2001); McConn et al., Plant J. 15, 521 (1998)) and are precursors of certain hormones and signaling molecules (Heller et al., Drugs 55, 487 (1998); Creelman et al., Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 355 (1997)). Known pathways of PUFA synthesis involve the processing of saturated 16:0 or 18:0 fatty acids (the abbreviation X:Y indicates an acyl group containing X carbon atoms and Y cis double bonds; double-bond positions of PUFAs are indicated relative to the methyl carbon of the fatty acid chain (ω3 or ω6) with systematic methylene interruption of the double bonds) derived from fatty acid synthase (FAS) by elongation and aerobic desaturation reactions (Sprecher, Curr. Opin. Clin. Nutr. Metab. Care 2, 135 (1999); Parker-Barnes et al., Proc. Natl. Acad. Sci. USA 97, 8284 (2000); Shanklin et al., Annu. Rev. Plant Physiol. Plant Nol. Biol. 49, 611 (1998)). Starting from acetyl-CoA, the synthesis of DHA requires approximately 30 distinct enzyme activities and nearly 70 reactions including the four repetitive steps of the fatty acid synthesis cycle. Polyketide synthases (PKSs) carry out some of the same reactions as FAS (Hopwood et al., Annu. Rev. Genet. 24, 37 (1990); Bentley et al., Annu. Rev. Microbiol. 53, 411 (1999)) and use the same small protein (or domain), acyl carrier protein (ACP), as a covalent attachment site for the growing carbon chain. However, in these enzyme systems, the complete cycle of reduction, dehydration and reduction seen in FAS is often abbreviated so that a highly derivatized carbon chain is produced, typically containing many keto- and hydroxy-groups as well as carbon-carbon double bonds in the trans configuration. The linear products of PKSs are often cyclized to form complex biochemicals that include antibiotics and many other secondary products (Hopwood et al., (1990) supra; Bentley et al., (1999), supra; Keating et al., Curr. Opin. Chem. Biol. 3, 598 (1999)).
Very long chain PUFAs such as docosahexaenoic acid (DHA; 22:6ω3) and eicosapentaenoic acid (EPA; 20:5ω3) have been reported from several species of marine bacteria, including Shewanella sp (Nichols et al., Curr. Op. Biotechnol. 10, 240 (1999); Yazawa, Lipids 31, S (1996); DeLong et al., Appl. Environ. Microbiol. 51, 730 (1986)). Analysis of a genomic fragment (cloned as plasmid pEPA) from Shewanella sp. strain SCRC2738 led to the identification of five open reading frames (Orfs), totaling 20 Kb, that are necessary and sufficient for EPA production in E. coli (Yazawa, (1996), supra). Several of the predicted protein domains were homologues of FAS enzymes, while other regions showed no homology to proteins of known function. On the basis of these observations and biochemical studies, it was suggested that PUFA synthesis in Shewanella involved the elongation of 16- or 18-carbon fatty acids produced by FAS and the insertion of double bonds by undefined aerobic desaturases (Watanabe et al., J. Biochem. 122, 467 (1997)). The recognition that this hypothesis was incorrect began with a reexamination of the protein sequences encoded by the five Shewanella Orfs. At least 11 regions within the five Orfs were identifiable as putative enzyme domains (See Metz et al., Science 293:290-293 (2001)). When compared with sequences in the gene databases, seven of these were more strongly related to PKS proteins than to FAS proteins. Included in this group were domains putatively encoding malonyl-CoA:ACP acyltransferase (MAT), 3-ketoacyl-ACP synthase (KS), 3-ketoacyl-ACP reductase (KR), acyltransferase (AT), phosphopantetheine transferase, chain length (or chain initiation) factor (CLF) and a highly unusual cluster of six ACP domains (i.e., the presence of more than two clustered ACP domains has not previously been reported in PKS or FAS sequences). However, three regions were more highly homologous to bacterial FAS proteins. One of these was similar to the newly-described Triclosan-resistant enoyl reductase (ER) from Streptococcus pneumoniae (Heath et al., Nature 406, 145 (2000)); comparison of ORF8 peptide with the S. pneumoniae enoyl reductase using the LALIGN program (matrix, BLOSUM50; gap opening penalty, −10; elongation penalty −1) indicated 49% similarity over a 386aa overlap). Two regions were homologues of the E. coli FAS protein encoded by fabA, which catalyzes the synthesis of trans-2-decenoyl-ACP and the reversible isomerization of this product to cis-3-decenoyl-ACP (Heath et al., J. Biol. Chem., 271, 27795 (1996)). On this basis, it seemed likely that at least some of the double bonds in EPA from Shewanella are introduced by a dehydrase-isomerase mechanism catalyzed by the FabA-like domains in Orf7.
Anaerobically-grown E. coli cells harboring the pEPA plasmid accumulated EPA to the same levels as aerobic cultures (Metz et al., 2001, supra), indicating that an oxygen-dependent desaturase is not involved in EPA synthesis. When pEPA was introduced into a fabB− mutant of E. coli, which is unable to synthesize monounsaturated fatty acids and requires unsaturated fatty acids for growth, the resulting cells lost their fatty acid auxotrophy. They also accumulated much higher levels of EPA than other pEPA-containing strains, suggesting that EPA competes with endogenously produced monounsaturated fatty acids for transfer to glycerolipids. When pEPA-containing E. coli cells were grown in the presence of [13C]-acetate, the data from 13C-NMR analysis of purified EPA from the cells confirmed the identity of EPA and provided evidence that this fatty acid was synthesized from acetyl-CoA and malonyl-CoA (See Metz et al., 2001, supra). A cell-free homogenate from pEPA-containing fabB− cells synthesized both EPA and saturated fatty acids from [14C]-malonyl-CoA. When the homogenate was separated into a 200,000×g high-speed pellet and a membrane-free supernatant fraction, saturated fatty acid synthesis was confined to the supernatant, consistent with the soluble nature of the Type II FAS enzymes (Magnuson et al., Microbiol. Rev. 57, 522 (1993)). Synthesis of EPA was found only in the high-speed pellet fraction, indicating that EPA synthesis can occur without reliance on enzymes of the E. coli FAS or on soluble intermediates (such as 16:0-ACP) from the cytoplasmic fraction. Since the proteins encoded by the Shewanella EPA genes are not particularly hydrophobic, restriction of EPA synthesis activity to this fraction may reflect a requirement for a membrane-associated acyl acceptor molecule. Additionally, in contrast to the E. coli FAS, EPA synthesis is specifically NADPH-dependent and does not require NADH. All these results are consistent with the pEPA genes encoding a multifunctional PKS that acts independently of FAS, elongase, and desaturase activities to synthesize EPA directly. It is likely that the PKS pathway for PUFA synthesis that has been identified in Shewanella is widespread in marine bacteria. Genes with high homology to the Shewanella gene cluster have been identified in Photobacterium profundum (Allen et al., Appli. Environ. Microbiol. 65:1710 (1999)) and in Moritella marina (Vibrio marinus) (Tanaka et al., Biotechnol. Lett. 21:939 (1999)).
The biochemical and molecular-genetic analyses performed with Shewanella provide compelling evidence for polyketide synthases that are capable of synthesizing PUFAs from malonyl-CoA. A complete scheme for synthesis of EPA by the Shewanella PKS has been proposed. The identification of protein domains homologous to the E. coli FabA protein, and the observation that bacterial EPA synthesis occurs anaerobically, provide evidence for one mechanism wherein the insertion of cis double bonds occurs through the action of a bifunctional dehydratase/2-trans, 3-cis isomerase (DH/2,3I). In E. coli, condensation of the 3-cis acyl intermediate with malonyl-ACP requires a particular ketoacyl-ACP synthase and this may provide a rationale for the presence of two KS in the Shewanella gene cluster (in Orf5 and Orf7). However, the PKS cycle extends the chain in two-carbon increments while the double bonds in the EPA product occur at every third carbon. This disjunction can be solved if the double bonds at C-14 and C-8 of EPA are generated by 2-trans, 2-cis isomerization (DH/2,2I) followed by incorporation of the cis double bond into the elongating fatty acid chain. The enzymatic conversion of a trans double bond to the cis configuration without bond migration is known to occur, for example, in the synthesis of 11-cis-retinal in the retinoid cycle (Jang et al., J. Biol. Chem. 275, 28128 (2000)). Although such an enzyme function has not yet been identified in the Shewanella PKS, it may reside in one of the unassigned protein domains.
The PKS pathways for PUFA synthesis in Shewanella and another marine bacteria, Vibrio marinus, are described in detail in U.S. Pat. No. 6,140,486 (issued from U.S. application Ser. No. 09/090,793, filed Jun. 4, 1998, entitled “Production of Polyunsaturated Fatty Acids by Expression of Polyketide-like Synthesis Genes in Plants”, which is incorporated herein by reference in its entirety).
Polyunsaturated fatty acids (PUFAs) are considered to be useful for nutritional, pharmaceutical, industrial, and other purposes. An expansive supply of PUFAs from natural sources and from chemical synthesis are not sufficient for commercial needs. Because a number of separate desaturase and elongase enzymes are required for fatty acid synthesis from linoleic acid (LA, 18:2 Δ9, 12), common in most plant species, to the more saturated and longer chain PUFAs, engineering plant host cells for the expression of PUFAs such as EPA and DHA may require expression of five or six separate enzyme activities to achieve expression, at least for EPA and DHA. Additionally, for production of useable quantities of such PUFAs, additional engineering efforts may be required, for instance the down regulation of enzymes competing for substrate, engineering of higher enzyme activities such as by mutagenesis or targeting of enzymes to plastid organelles. Therefore it is of interest to obtain genetic material involved in PUFA biosynthesis from species that naturally produce these fatty acids and to express the isolated material alone or in combination in a heterologous system which can be manipulated to allow production of commercial quantities of PUFAs.
The discovery of a PUFA PKS system in marine bacteria such as Shewanella and Vibrio marinus (see U.S. Pat. No. 6,140,486, ibid.) provides a resource for new methods of commercial PUFA production. However, these marine bacteria have limitations which will ultimately restrict their usefulness on a commercial level. First, although U.S. Pat. No. 6,140,486 discloses that the marine bacteria PUFA PKS systems can be used to genetically modify plants, the marine bacteria naturally live and grow in cold marine environments and the enzyme systems of these bacteria do not function well above 30° C. In contrast, many crop plants, which are attractive targets for genetic manipulation using the PUFA PKS system, have normal growth conditions at temperatures above 30° C. and ranging to higher than 40° C. Therefore, the marine bacteria PUFA PKS system is not predicted to be readily adaptable to plant expression under normal growth conditions. Moreover, the marine bacteria PUFA PKS genes, being from a bacterial source, may not be compatible with the genomes of eukaryotic host cells, or at least may require significant adaptation to work in eukaryotic hosts. Additionally, the known marine bacteria PUFA PKS systems do not directly produce triglycerides, whereas direct production of triglycerides would be desirable because triglycerides are a lipid storage product in microorganisms and as a result can be accumulated at very high levels (e.g. up to 80-85% of cell weight) in microbial/plant cells (as opposed to a “structural” lipid product (e.g. phospholipids) which can generally only accumulate at low levels (e.g. less than 10-15% of cell weight at maximum)).
Therefore, there is a need in the art for other PUFA PKS systems having greater flexibility for commercial use.